For years, Nvidia has been the silent power behind the world’s most famous AI models. But now the company is quietly stepping back from deeper financial ties with the creators of ChatGPT and Claude — a move that could reshape the balance of power in the AI industry.
For the first time in modern tech history, an AI company is pushing back against direct pressure from U.S. defence and intelligence agencies.
At the centre of it: Anthropic — and a standoff that could define how AI will be used in national security for the next decade.
For years, the Pentagon has quietly expanded its access demands on tech companies.
Now, with frontier AI systems entering defence workflows, the friction has reached a new level — and Anthropic is at the centre of it.
What started as a challenger to the dominant AI labs has turned into one of the biggest waves in the tech world: Anthropic is dominating headlines with new models, massive funding talks, major enterprise deals — and yes, even stock market tremors sparked by the impact of its AI tools. In just weeks, Anthropic has gone from “cool AI startup” to a market mover with a bold playbook that’s disrupting the industry.
AI search startup Perplexity has signed a massive $750 million cloud computing agreement with Microsoft, a three-year pact that could reshape how AI companies choose their infrastructure and where frontier models live. But beyond the dollar figure, this deal reflects a broader shift in the AI ecosystem — one where flexibility, multi-cloud strategies, and competitive positioning matter as much as raw compute capacity.
user
thebohemianbliss